We have now completed our studies on the nature of the nephritogenic protein using the sera and renal biopsies from patients with acute post streptococcal glomerulonephritis (APSGN). All nephritogenic strains produce a protein called nephritis plasmin binding protein (NPBP) which was isolated, sequenced and found to be identical with streptococcal proteinase or streptococcal pyrogenic exotoxin B (SPEB). Using antibodies prepared against recombinant SPEB, 65% of 20 APSGN biopsies were found to contain the antigen in the glomerulus while only 4% of 25 non APSGN biopsies were positive. None of the biopsies were positive for streptokinase, another streptococcal antigen possibly associated with APSGN. APSGN sera reacted preferentially with this antigen when compared to sera from patients with either rheumatic fever or uncomplicated streptococcal infections and normal controls. Furthermore, serial serum studies of APSGN patients revealed that while the titers decreased over one year they never returned to baseline values suggesting a possible protective effect against the known fact that recurrences of APSGN are extremely rare. A manuscript detailing these findings has been submitted to Kidney International. Studies involving the rheumatic fever marker D8/17 have proceeded along two areas of investigation. The first is concerned with the predictive role of the marker for disease susceptibility. Approximately 3,000 children ages 5-10 years who come from high risk areas of rheumatic fever in Mexico City have been tested for the marker. Seven percent of these unaffected children are positive for the marker. All children are being followed over time for the appearance of rheumatic fever. If our hypothesis is correct, only those positive for the D8/17 marker will be susceptible. The second area of investigation of the marker was unexpected. In collaboration with a group from Child Psychiatry at NIH under the direction of Dr. Sue Swedo, we have examined the presence or absence of this marker in a group of 23 children (and appropriate controls) with obsessive-compulsive disorders (OCD). In a double blind test the marker correctly identified ninety percent of the OCD patients compared to the expected 7% controls. In view of the known cross-reactions between streptococcal antigens and caudate nucleus cells, these studies suggest that other brain-streptococcal cross reactive antigens may be involved in the OCD syndrome. A manuscript detailing these findings has been sent to the J. Child Psychaitry. Our more basic approach to the exact nature of the D8/17 antigen and caudate binding antigen in the sera of Sydenham's chorea are being pursued along two main lines of investigation. Concerning the D8/17 antigen our main problem in identifying this antigen has been the fact that the antibody is the IgM class. Thus non-specific binding of other proteins has resulted in identification of a number of bands. Secondly, this antibody does recognize other antigens expressing a coil-coiled structure. We have recently isolated a IgG clone of the D8/17 antibody and hope this will have the same specificity as the IgM molecule. We are also using the chemiluminescence technique (quite sensitive) to further identify the putative antigen. Concerning the Sydenham's caudate antigen we are screening sera from these patients both on a cDNA library of human caudate as well as immunoblots of human and mouse caudate specimens. In this respect we have identified two bands of 82 and 66 Kda that appear promising.